TITLE {-}

Authors: Danielle C. Claar \(^1\), Kristina L. Tietjen \(^1\), Ruth D. Gates \(^2\), Julia K. Baum \(^1\)

Institute: \(^1\) Department of Biology, University of Victoria, PO BOX 1700 Station CSC, Victoria, British Columbia, V8W 2Y2, Canada \(^2\) Hawaii Institute of Marine Biology, 46-007 Lilipuna Road, Kaneohe, HI 96744, USA

Corresponding Author: Danielle C. Claar, Tel: (208) 250-0161, Email: dclaar@uvic.ca

Keywords: coral bleaching, heat stress, climate change, Symbiodinium

Summary

Coral reefs, which already live on the edge of their thermal tolerance¹, are under acute threat from ocean warming^2–4. Corals live in symbiosis with an extraordinarily diverse genus of photosynthetic dinoflagellates (Symbiodinium spp.;^5,6). The symbiotic association and diversity of various taxa of Symbiodinium can be flexible over time^7,8, but see^9,10, and individual Symbidoinium taxa can range from parasites to mutualists in their interaction with their coral host¹¹. Warming causes the breakdown of coral symbiosis, causing coral “bleaching” when symbionts are expelled and the white coral skeleton is visible through the coral tissue¹². Coral bleaching can lead to mortality, although corals can regain their symbionts after heat stress has abated^13,14. The 2015/16 El Niño is the worst pulse warming event on record in terms of severity and longevity^???,15, yet despite massive coral mortality, some corals show resilience to this extreme event (???). Here, we track coral symbioses and survival at the epicenter of this bleaching event (Kiritimati, Central Pacific), and show, contrary to our current paradigm of coral bleaching and recovery dynamics, that some corals have the capacity to re-establish symbiosis before heat stress subsides. Furthermore, we demonstrate potential mechanisms for coral survival and recovery, including the lack of preferential symbiont expulsion, and the effect of local human disturbance on pre-bleaching symbiont community structure and the probability of coral survival. Together, these results show the potential for reef corals to survive extreme warming events, providing tentative hope for the survival of corals in the Anthropocene.

Main Text

Global coral bleaching is increasing, and the 2014-2017 event caused a catastrophic loss of corals around the globe. There was up to 95% mortality in some regions during the 1997/1998 El Niño event (Glynn 1993). The 2014-2017 global coral bleaching event caused coral bleaching across the world’s oceans (Eakin 2016, Normile 2016), with up to 75% bleaching on some reefs in Hawaii, and at least some level of bleaching across 93% of the Great Barrier Reef (Minton et al 2015, GBRMPA 2016). The 2015-2016 El Niño, superimposed on nearly-ubiquitous tropical ocean warming, instigated the third global coral bleaching event (???).

The symbiosis between coral and their single-celled dinoflagellate symbionts, Symbiodinium, is the foundation of reef ecosystems, and a critical element of reef resilience [16; Muller-Parker2015-sd]. The coral holobiont responds to environmental conditions, and is the unit that interacts with the broader reef community¹⁷, supporting reef diversity and function across taxa (or at a global scale?). There is much genetic, functional, and response diversity within the Symbiodinium genus. Symbiodinium types, considered putative species¹⁸, have distinct geographic distributions, host associations, and environmental optima¹⁹. There are functional differences between Symbiodinium clades²⁰, and Symbiodinium associations can range from mutualistic to neutral to parasitic based on Symbiodinium type as well as environmental conditions¹¹. Recent advances in next-generation sequencing techniques have revealed cryptic genetic diversity within symbiotic Symbiodinium^21–23, and has allowed for long-term genetic and ecological comparisons of symbiont community structure²⁴.

The adaptive bleaching hypothesis suggests that corals bleach in order to expel environmentally sub-optimal symbionts, followed by switching (picking up new symbionts from the environment) or shuffling (an internal change in dominant symbiont type or overall symbiont community structure)^7,25–27. There is ample evidence for Symbiodinium shuffling (Rowan 2004), and a recent study showed evidence for Symbiodinium switching²⁸. However, what remains unclear is if and how frequently bleaching events can actually be considered adaptive. Changes in photosynthetic efficiency during bleaching as well as bleaching resistance have been shown to correspond to distinct Symbiodinium phylotypes²⁹. Clade D Symbiodinium are proported to have an enhanced thermal tolerance³⁰, and repopulation of a coral host with clade D symbionts after a bleaching event is proposed to be a survival mechanism^31–33. A history of thermal stress increased the prevalence of clade D Symbiodinium in a generalist coral species, but did not instigate similar changes in two specialist coral species³⁴. Although the prevalence of clade D Symbiodinium increases during thermal stress and may increase thermal tolerance³¹, corals that house clade D symbionts may have slower growth rates⁸ or lower energy storage³⁵. Furthermore, functional differences exist not only at the clade level, but are present among types within a single clade³⁶.

The current paradigm of coral bleaching and resilience is that as environmental stress (such as warming) increases, corals begin to lose their obligate symbionts (Symbiodinium) and “bleach”^13,37. Thermal stress is the primary cause for coral bleaching, and extreme or long-lasting warming causes a complete breakdown of the coral symbioses, leading to expulsion of all (or nearly all) Symbiodinium from the coral host tissue, leading to mortality³⁸. Thermal stress can be exacerbated by other environmental stressors (Cooper et al 2011, Béraud et al 2013, Maina et al 2008). During bleaching, there is a window for recovery, that is, a certain amount of time during which the warming must cease and conditions must return to normal so that the coral can regain its symbionts. If the window for recovery passes without amelioration of the environmental conditions, the coral will starve and die. (Cunning et al 2016, Putnam et al 2017). Here we show that despite unprecedented heat stress, some corals exhibited resilience and survived. Survival through such an extreme heat event provides an exceptional opportunity to understand how some corals can withstand intense heat stress, and how corals in general might survive long-term warming. Remarkably, we find that some coral colonies were able to survive this prolonged heat stress by regaining their symbionts while temperatures were still elevated.

Our study location, Kiritimati Atoll (Christmas Island, Kiribati, Central Equatorial Pacific, Coordinates: 2, -157.4), was at the epicenter of this extreme El Niño event. Thermal anomalies were severe on Kiritimati, rapidly exceeding NOAA Coral Reef Watch’s Coral Bleaching Alert Level 1 and Alert Level 2 thresholds, reaching an unprecedented (39) 25.7 DHW over a year-long bleaching event, demolishing most of the reef (???). Despite these staggering losses, some corals have the capacity to be resilient to these increasingly frequent mass-bleaching events (Hughes et al 2017). Here, we assess coral symbiosis and survival during the massive 2015/2016 El Niño event. We tagged, sampled, and photographed the same coral colonies before, during, and immediately after the El Niño event. We assessed bleaching condition and survival for each coral colony, and used Illumina MiSeq ITS2 amplicon sequencing and 97% de novo OTU clustering to evaluate changes in Symbiodinium community structure. To investigate mechanisms underlying the ability of these corals to not only survive a year of continuous heat stress, but to recover in the interim, we assessed the relationship between human disturbance, pre-bleaching Symbiodinium community structure, and coral survival, as well as the timing of Symbiodinium community shifts throughout this El Niño event.

We document, for the first time, corals that were able to visually recover from bleaching, and to regain their Symbiodinium communities during the course of an extreme heat stress event. These corals (family Faviidae; Platygyra sp. and Favites sp.) were bleached within two months of the onset of warming, but had visibly recovered after 10 consecutive months of intense warming (Fig. 1).

Figure 1. Top: Heat stress, expressed as Degree Heating Weeks (DHW), on Kiritimati Island over the course of the 2015-2016 El Niño event. Corals are sensitive to temperatures warmer than 1&degC above their normal highest summertime mean sea surface temperature (SST), known as the bleaching threshold. DHW shows how much heat stress has accumulated in an area over the past twelve weeks by adding up any temperature exceeding the bleaching threshold during that time period. Horizontal lines show expected bleaching severity levels: 4 &degC (yellow line), NOAA Coral Reef Watch (CRW) Bleaching Alert Level 1, where significant coral bleaching is likely; 8 &degC (light orange line), NOAA CRW Bleaching Alert Level 2, where widespread bleaching and mortality may occur; 12 &degC (dark orange line), mass coral mortality (Hoegh-Guldberg 2011); 24 &degC (dark red line) “not experienced by reefs yet” (Hoegh-Guldberg 2011). Solid black line indicates in situ calculated DHW, and fill colors correspond to bleaching temperature thresholds. The dashed vertical gray line show the five sampling time points. Bottom: Photographs of a single tagged Platygyra coral colony (#99) at five sampling time points, showing the initially healthy colony bleached after two months of heat stress, ‘recovered’ to a normal brown colour after ten months of heat stress, and still alive six months post heat stress (Note: no photo is shown from the May 2015 sampling time point). NOTE: Add ‘degree C-weeks’ (which is the units for DHW to the y-axis of the plots; use degree symbol instead of the word degree)

The “transient microbiome” (of rare Symbiodinium types) assembled by environmental anomalies can undergo rapid changes (Putnam et al 2017), providing symbiotic stochasticity which may build or weaken a coral’s capacity for resilience. Corals commonly host background Symbiodinium types in low levels (Correa et al 2009), but sub-dominant Symbiodinium communities are often unstable (Coffroth et al 2010). The importance of rare Symbiodinium types is currently under debate, and these rare types may be commensal (symbionts that pass through coral’s holobiont with no harm or gain for either partner), parasitic (“cheaters”, or symbionts that take more than they give), or mutualistic (symbionts which support host function) (Parkinson et al 2015). Some research suggests that low-abundance Symbiodinium types have minimal functional significance to corals (Lee et al 2016), while other evidence supports the idea that the rare Symbiodinium biosphere is important for corals’ response to climate change (Boulotte et al 2016) and that, shifts in Symbiodinium community diversity may have a larger influence on coral resilience than the evolution of symbiont thermal tolerance (Baskett et al 2010). In other systems, rare microbial species have been demonstrated to be disproportionally important to maintaining functional processes during environmental change (Shade et al 2014). We show that after two months of heat stress, fully-bleached corals retained approximately the same Symbiodinium community as they had before the bleaching event. This suggests that a wholesale breakdown of symbiosis occurred in bleached corals during this event, indicating a lack of preferential symbiont expulsion or exodus. Furthermore, we demonstrate that symbionts present in even very low abundances can play a critical role in coral survival and recovery. Some coral colonies recovered symbiosis with Symbiodinium types that were present in only a negligible amount before the bleaching event.

[Figure 2. Symbiodinium community composition at each of five time points, showing the shift in dominance from clade C to clade D over the course of the 2015-2016 El Ni̱o event, for A. the entire pool of tagged Platygyra coral colonies at each of the five sampling time points (n= X - Y colonies per time point), B. the same individual tagged Platygyra coral colony (#99). РThis figure illustrates non-preferential explusion of clades c and d; and that symbiodinium that ar einitially extremely rare can play a critical role in coral resilience to heat stress.]

Local protection is critical for coral symbiosis and survival. We show that corals living at different levels of local human disturbance had distinct symbiont communities that corresponded tightly to survivorship. This is in contrast to a recent study which concluded that particulate and dissolved nutrients do not reduce coral health at a colony scale (Rocker et al 2017). However, there is increasing evidence for local adaptation in corals (Howells et al 2012, Logan et al 2013, Dixon et al 2015). Our results suggest that some Kiritimati coral species may have the capacity to experience evolutionary rescue, defined as adaptation at a rate that allows an endangered population to survive the rate of environmental change (Orr & Unkless 2014, Carlson 2014). Our results suggest that the capacity for evolutionary rescue is tangibly related to local reef protection. Although massive bleaching events like this one will likely continue to cause catastrophic damage to coral reefs worldwide, mitigating local human disturbance can potentially help protect some coral species against a modest amount of ocean warming.

[Figure 3. A. [Danielle to write this one: - will be the Constrained ordination plot showing groupings of Symbiodinium communities from individual Platygyra colonies, grouping into two distinct areas according to level of local disturbance….]; B. Bar plots showing Symbiodinium community composition for individual Platygya colonies at a single time point prior to the heat stress, from sites with high (top) and low (bottom) levels of local disturbance levels.] Figure 4. Potentially the symbiodinium network plot. Figure 5. Potentially the rank abundance plot for Platy…..

Methods

The Methods section should be written as concisely as possible but should contain all elements necessary to allow interpretation and replication of the results. As a guideline, Methods sections typically do not exceed 3,000 words. Detailed descriptions of methods already published should be avoided; a reference number can be provided to save space, with any new addition or variation stated. The Methods section should be subdivided by short bold headings referring to methods used and we encourage the inclusion of specific subsections for statistics, reagents and animal models. If further references are included in this section, the numbering should continue from the end of the last reference number in the rest of the paper and the list should accompany the additional Methods at the end of the paper. The Methods section cannot contain figures or tables (essential display items should be included in the Extended Data).

0.1 Field Sampling

Kiritimati Atoll (Christmas Island), Kiribati is located in the Central Equatorial Pacific (1.9N 157W), at the center of the El Niño 3.4 region (a region which is used to quantify El Niño presence and strength). During the 2015/2016 El Niño event, Kiritimati experienced 10 months of sustained temperature stress, causing a mass bleaching and mortality event (???).

0.2 Temperature quantification

Temperature loggers (Sea-Bird 56) were deployed around the island at 10-12m depth from 2011-2016 to measure in situ thermal stress.

0.3 Coral Tagging and sampling

In August/September 2014 colonies of Platygyra sp. and Favites pentagona were tagged along a 60m transect at 10-12m depth at 15 different sites around the atoll. A photo was taken of each coral to record colony measurments and characteristics (i.e. bleaching). The tagged coral colonies were resampled twice more before (January/February 2015, April/May 2015), once during (July 2015), and once near the end (March 2016) of the El Niño warming. Some tagged coral colonies were lost due to storm damage, and new coral colonies were tagged to replenish the total number of surveyed colonies. Not all sites were visited during all field seasons, and some site surveys were only partially completed during some field seasons due to inclement weather conditions.

Corals were sampled by… processed like this… Coral tissue samples were preserved in Guanidinium buffer (50% w/v guanidinium isothiocyanate; 50 mM Tris pH 7.6; 10 µM EDTA; 4.2% w/v sarkosyl; 2.1% v/v β-mercaptoethanol) and stored at 4 degrees until extraction.

0.4 Pre-processing and sequencing

DNA extraction was performed using a guanidinium-based extraction protocol [14; Cunning2017-sc; Cunning2015-mt] with the modification that the DNA pellet was washed with 70% ethanol three times rather than once.

Library Prep- Amy’s method of library prep, include cleanup, Illumina Sequencing information (barcodes, etc) for sequencing of ITS2 amplicons (‘itsD’ and ‘its2rev2’ primers from¹⁴ Illumina MiSeq platform with 2×300 paired-end read chemistry. HIMB - Amy

ITS2 region - it’s annoying, but it’s the best we’ve got right now smith et al 2017

A total of 289 samples were prepared for sequencing, and XXXX of these samples were successfully amplified, sequenced, and used in downstream analyses.

0.5 Bioinformatics

We conducted quality filtering of raw reads (in .fastq format) first using iu-filter-quality-bokulich implemented in Illumina-Utils [40; Eren2013-yg], followed by paired-end sequence merging via iu-merge-pairs (also in Illumina-Utils, [Eren2013-yg]), with a maximum mismatch of three bases between the forward and reverse reads. After quality filtering, sequence processing and identification was performed following all specifications of⁴¹; chimeric sequences were removed, primers were trimmed, sequences from each sample were clustered independently at 97% similarity using UCLUST⁴² implemented in QIIME⁴³ and resulting OTUs were collapsed at 100% identity across samples, sequences were aligned using the Needleman-Wunsch global alignment algorithm (Biostrings package,⁴⁴) in R⁴⁵, and sequences were named using a reference database.

The Phyloseq package⁴⁶ in R was used to store and analyze OTU tables, taxonomic information, and sample metadata. The phyloseq object was filtered to remove OTUs observed <10 times, which removed more than half of observed otus (n=83 OTUs removed and n=81 kept, including n=10 doubletons). The phyloseq object was further filtered to remove samples with very low sequence abundance (<200 sequences, n=27 samples removed and n=262 kept).

In 262 coral samples, we found XXXX sequences after quality filtering. clade abundances here

0.6 Statistical Analysis

Code will be avaible on git hub

References

1. Sampayo, E. M. et al. Coral symbioses under prolonged environmental change: Living near tolerance range limits. Sci. Rep. 6, 36271 (2016).

2. Hughes, T. P. et al. Climate change, human impacts, and the resilience of coral reefs. Science 301, 929–933 (2003).

3. Hoegh-Guldberg, O. et al. Coral reefs under rapid climate change and ocean acidification. Science 318, 1737–1742 (2007).

4. Baker, A. C., Glynn, P. W. & Riegl, B. Climate change and coral reef bleaching: An ecological assessment of long-term impacts, recovery trends and future outlook. Estuar. Coast. Shelf Sci. 80, 435–471 (2008).

5. Muscatine, L. & Porter, J. W. Reef corals: Mutualistic symbioses adapted to Nutrient-Poor environments. Bioscience 27, 454–460 (1977).

6. Rowan, R. & Powers, D. A. Ribosomal RNA sequences and the diversity of symbiotic dinoflagellates (zooxanthellae). Proc. Natl. Acad. Sci. U. S. A. 89, 3639–3643 (1992).

7. Baker, A. C. Flexibility and specificity in Coral-Algal symbiosis: Diversity, ecology, and biogeography of symbiodinium. Annu. Rev. Ecol. Evol. Syst. 34, 661–689 (2003).

8. Little, A. F., Oppen, M. J. H. van & Willis, B. L. Flexibility in algal endosymbioses shapes growth in reef corals. Science 304, 1492–1494 (2004).

9. Thornhill, D. J., LaJeunesse, T. C., Kemp, D. W., Fitt, W. K. & Schmidt, G. W. Multi-year, seasonal genotypic surveys of coral-algal symbioses reveal prevalent stability or post-bleaching reversion. Mar. Biol. 148, 711–722 (2006).

10. Thornhill, D. J., Xiang, Y., Fitt, W. K. & Santos, S. R. Reef endemism, host specificity and temporal stability in populations of symbiotic dinoflagellates from two ecologically dominant caribbean corals. PLoS One 4, e6262 (2009).

11. Lesser, M. P., Stat, M. & Gates, R. D. The endosymbiotic dinoflagellates (symbiodinium sp.) of corals are parasites and mutualists. Coral Reefs 32, 603–611 (2013).

12. Brown, B. E. Coral bleaching: Causes and consequences. Coral Reefs 16, S129–S138 (1997).

13. Douglas, A. E. Coral bleaching—-how and why? Mar. Pollut. Bull. 46, 385–392 (2003).

14. Stat, M., Loh, W. K. W., LaJeunesse, T. C., Hoegh-Guldberg, O. & Carter, D. A. Stability of coral–endosymbiont associations during and after a thermal stress event in the southern great barrier reef. Coral Reefs 28, 709–713 (2009).

15. Heron, S. F., Maynard, J. A., Hooidonk, R. van & Eakin, C. M. Warming trends and bleaching stress of the world’s coral reefs 1985-2012. Sci. Rep. 6, 38402 (2016).

16. Oppen, M. J. H. van & Gates, R. D. Conservation genetics and the resilience of reef-building corals. Mol. Ecol. 15, 3863–3883 (2006).

17. Gates, R. D. & Ainsworth, T. D. The nature and taxonomic composition of coral symbiomes as drivers of performance limits in scleractinian corals. J. Exp. Mar. Bio. Ecol. 408, 94–101 (2011).

18. Pochon, X. & Gates, R. D. A new symbiodinium clade (dinophyceae) from soritid foraminifera in hawai’i. Mol. Phylogenet. Evol. 56, 492–497 (2010).

19. Fabina, N. S., Putnam, H. M., Franklin, E. C., Stat, M. & Gates, R. D. Transmission mode predicts specificity and interaction patterns in coral-symbiodinium networks. PLoS One 7, e44970 (2012).

20. Stat, M., Morris, E. & Gates, R. D. Functional diversity in coral–dinoflagellate symbiosis. Proceedings of the National Academy of Sciences 105, 9256–9261 (2008).

21. Quigley, K. M. et al. Deep-sequencing method for quantifying background abundances of symbiodinium types: Exploring the rare symbiodinium biosphere in reef-building corals. PLoS One 9, e94297 (2014).

22. Arif, C. et al. Assessing symbiodinium diversity in scleractinian corals via next-generation sequencing-based genotyping of the ITS2 rDNA region. Mol. Ecol. 23, 4418–4433 (2014).

23. Green, E. A., Davies, S. W., Matz, M. V. & Medina, M. Quantifying cryptic symbiodinium diversity within orbicella faveolata and orbicella franksi at the flower garden banks, gulf of mexico. PeerJ 2, e386 (2014).

24. Edmunds, P. J. et al. Long-term changes in symbiodinium communities in orbicella annularis in st. john, US virgin islands.

25. Buddemeier, R. W., Baker, A. C., Fautin, D. G. & Jacobs, J. R. The adaptive hypothesis of bleaching. Coral health and disease 427444, (2004).

26. Buddemeier, R. W. & Fautin, D. G. Coral bleaching as an adaptive mechanism. Bioscience 43, 320–326 (1993).

27. Baker, A. C. Ecosystems: Reef corals bleach to survive change. Nature 411, 765–766 (2001).

28. Boulotte, N. M. et al. Exploring the symbiodinium rare biosphere provides evidence for symbiont switching in reef-building corals. ISME J. 10, 2693–2701 (2016).

29. Kemp, D. W., Hernandez-Pech, X., Iglesias-Prieto, R., Fitt, W. K. & Schmidt, G. W. Community dynamics and physiology of symbiodinium spp. before, during, and after a coral bleaching event. Limnol. Oceanogr. 59, 788–797 (2014).

30. Stat, M. & Gates, R. D. Clade D symbiodinium in scleractinian corals: A ‘nugget’ of hope, a selfish opportunist, an ominous sign, or all of the above? J. Mar. Biol. 2011, (2010).

31. Berkelmans, R. & Oppen, M. J. H. van. The role of zooxanthellae in the thermal tolerance of corals: A ’nugget of hope’ for coral reefs in an era of climate change. Proc. Biol. Sci. 273, 2305–2312 (2006).

32. Mieog, J. C. et al. Real-time PCR reveals a high incidence of symbiodinium clade D at low levels in four scleractinian corals across the great barrier reef: Implications for symbiont shuffling. Coral Reefs 26, 449 (2007).

33. Silverstein, R. N., Correa, A. M. S. & Baker, A. C. Specificity is rarely absolute in coral–algal symbiosis: Implications for coral response to climate change. Proceedings of the Royal Society of London B: Biological Sciences 279, 2609–2618 (2012).

34. Stat, M. et al. The distribution of the thermally tolerant symbiont lineage (symbiodinium clade d) in corals from hawaii: Correlations with host and the history of ocean thermal stress. Ecol. Evol. 3, 1317–1329 (2013).

35. Jones, A. M. & Berkelmans, R. Tradeoffs to thermal acclimation: Energetics and reproduction of a reef coral with heat tolerant symbiodinium Type-D. J. Mar. Biol. 2011, (2011).

36. Sampayo, E. M., Ridgway, T., Bongaerts, P. & Hoegh-Guldberg, O. Bleaching susceptibility and mortality of corals are determined by fine-scale differences in symbiont type. Proceedings of the National Academy of Sciences 105, 10444–10449 (2008).

37. Gates, R. D., Baghdasarian, G. & Muscatine, L. Temperature stress causes host cell detachment in symbiotic cnidarians: Implications for coral bleaching. Biol. Bull. 182, 324–332 (1992).

38. Hoegh-Guldberg, O. Climate change, coral bleaching and the future of the world’s coral reefs. Mar. Freshwater Res. 50, 839–866 (1999).

39. Hoegh-Guldberg, O. Coral reef ecosystems and anthropogenic climate change. Regional Environ. Change 11, 215–227 (2011).

40. Bokulich, N. A. et al. Quality-filtering vastly improves diversity estimates from illumina amplicon sequencing. Nat. Methods 10, 57–59 (2013).

41. Cunning, R., Gates, R. D. & Edmunds, P. J. Using high-throughput sequencing of ITS2 to describe symbiodinium metacommunities in st. john, US virgin islands. PeerJ 5, e3472 (2017).

42. Edgar, R. C. Search and clustering orders of magnitude faster than BLAST. Bioinformatics 26, 2460–2461 (2010).

43. Caporaso, J. G. et al. QIIME allows analysis of high-throughput community sequencing data. Nat. Methods 7, 335–336 (2010).

44. Pagès, H., Aboyoun, P., Gentleman, R. & DebRoy, S. Biostrings: String objects representing biological sequences, and matching algorithms. (2017).

45. R Development Core Team. R: A language and environment for statistical computing. (2008).

46. McMurdie, P. J. & Holmes, S. Phyloseq: An R package for reproducible interactive analysis and graphics of microbiome census data. PLoS One 8, e61217 (2013).